D-R12l 611 ATMOSPHERIC EFFECTS IN LOW EARTH ORBIT AND THE DMSP ESA iniOFFSET ANOMALY. (U) AEROSPACE CORP EL SEGUNDO CRCHEMISTRY AND PHYSICS LAB 6 S ARNOLD ET AL. 30 SEP 82

UNCLASSIFIED TR-0082(247B-26)-i SD-TR-82-Bi F/G 22/3, L

I 1lflfflflfflfllflfllflf

flfllfllflfflf ND

7.: 'IE

II~ ~~" II 5 3-

W0 Lo_

III 11.8

1111125 -6 ~* fll....

MICROCOPY RESOLUTION TEST CHARTNATIONAL SUREAU OF S1ANAN- 1963 A

REPORT SD-TR-82-81

(=

Atmospheric Effects in Low Earth Orbit andthe DMSP ESA Offset Anomaly

0. S. ARNOLD, R. R. HERM, and D. R. PEPLINSKIChemistry and Physics Laboratory

Laboratory OperationsThe Aerospace Corporation •

El Segundo, Calif. 90245

'S 30 September 1982

APPROVED FOR PUBLIC RELEASE: U IDISTRIBUTION UNLIMITED 'jT f. .

NOV 1 9 1982

I A

* g Prepared for

SPACE DIVISIONAIR FORCE SYSTEMS COMMAND

LA -- Los Angeles Air Forct StationP.O. Box 92960, Worldway Postal Center

2 Los Angeles, Calif. 90009

82

This report was submitted by The Aerospace Corporation, El Segundo, CA

90245, under Contract No. F04701-81-C-0082 with the Space Division, Deputy for

Technology, P.O. Box 92960, Worldway Postal Center, Los Angeles, CA 90009. It

was reviewed and approved for The Aerospacd Corporation by S. Feuerstein,

Director, Chemistry and Physics Laboratory. Lt G- A- Wc.17 gn/VT u

was the project officer for Technology.

This report has been reviewed by the Public Affairs Office (PAS) and is

releasable to the National Technical Information Service (NTIS). At NTIS, it

will be available to the general public, including foreign nations.

This technical report has been reviewed and is approved for publication.

Publication of this report does not constitute Air Force approval of the

* report's findings or conclusions. It is published only for the exchange and

stimulation of ideas.

Gregory A. Welz, 2nd Lt, USAF iiie H. Butler, Colonel, USAFProject Officer Director of Space Systems Planning

FOR THE COMMANDER

Norman W, Lee, Jr., Colonel, USA4FDeputy for Technology

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE tehn Date Entered)

REPORT DOCUMENTATION PAGE READ INSTRUCTIONSBEFORE COMPLETING FORM

I. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

SD-TR-82-81 M -- /4. TITLE (and Subtitle) 1S. TYPE OF REPORT & PERIOD COVERED

ATMOSPHERIC EFFECTS IN LOW EARTH ORBIT ANDTHE DMSP ESA OFFSET ANOMALY

6. PERFORMING ORG. REPORT NUMBER

[_ _ _ _ _ _TR-0082 (2478-20)-i7. AUTHOR(&) U. CONTRACT OR GRANT NUMBER(e)

Graham S. Arnold, Ronald R. Herm, andDaniel R. Peplinski F04701-81-C-0082

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT, TASK

AREA A WORK UNIT NUMBERSThe Aerospace CorporationEl Segundo, Calif. 90245

11. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

Space Division 30 September 1982Air Force Systems Command 13. NUMBER OF PAGES

Los Angeles, Calif. 90009 5814. MONITORING AGENCY NAME & ADDRESS(If different from Controlling Office) IS. SECURITY CLASS. (of thli report)

UnclassifiedISa. DECLASSIFICATION/DOWNGRADING

SCHEDULE

16. DISTRIBUTION STATEMENT (of thie Report)

Approved for public release; distribution unlimited

17. DISTRIBUTION STATiMENT (of the abstract entered in Block 20, If different from Report)

18. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reverse aide If neceesary and identlir a-

Atmospheric Effects DMSPAtomic Beams GermaniumBeam-Surface Chemistry Oxygen AtomsContamination RTV-615

20. ABSTRACT (Continue on reverse side it neceeeary and identify by block number)

A s2ice vehicle in Earth orbit experiences bombardment with fast (about 8 km /5s4-) atmospheric particles, the principal of which is atomic oxygen, byvirtue of its orbital velocity. Several satellite systems that have been ormay be adversely affected by atmospheric impact are described. The currentknowledge of the gas-surface interactions of atomic oxygen and the availableexperimental technology for laboratory simulation of atmospheric bombardmentin orbit are reviewed. The DMSP ESA offset anomaly is described and evidence -

FORM 17GD IcM,-I 1473 UNCLASSIFIED

SECURITY CLASSIFICATION OF THIS PAGE (iften Date Entered)

/

UNCLASSIFIEDSECURITY CLASSFICATION OF THIS PA@E(Who VW& af.m4Is. KeY WORDS (Contlua.d)

20. ABSTRACT (Continued)

- for linking this misbehavior to atmospheric oxygen atom impact is presentedand discussed. The results of a laboratory experiment to test this linkare presented. <-

UNCLASSIFIEDSECURITY CLASSIFICATION OF TIS PAGfWIM Dale 209e,00

S . . . . . . . . .

ACKNOWLEDGMENTS

The authors wish to thank L. T. Greenberg of Aerospace, G. Nilsen of

Optical Coating Laboratories, Inc., R. Predmore of Goddard Space Flight

Center, and K. A. Ward of Barnes Engineering Co., for supplying unpublished

data and analyses which were of great use in the preparation of this report.

! •7

11'3

CONTENTS

.ACKNOWLEDGMENTS .......................................................... 9

Io INTRODUCTIONoo oo....oo.o~ooeooooe*oooooo 9

~~II. BACKGROUNDo ....................................... 11

A. Atmospheric Composition .......................... ....... 11

B. Atmospheric Effects on Orbit................................ 14

C. Oxygen Atom Surface Chemistry ................................ 19

D. Experimental Technology........................................ 23

III. DEFENSE METEOROLOGICAL SATELLITE EARTH SENSORASSEMBLY OFFSET ANOMALY................... .......... ............... 27

A. Earth Sensor Assembly.......................................... 27

B. ESA Degradation .... *.......... o................. ......... 30

C. Nimbus 6 and 7 ERB Instrument .................................. 32

D. Degradation Mechanisms ............... ....... ............. 33

E. DMSP ESA Offset Anomaly Laboratory Experiment .............. 4.. 39

IV. EXPERIMENTAL APPARATUS ..... o....... .ee ......... .......... ....... 41

A. Vacuum System........................... 41

B. Atomic Beam Source. .. .............. ..... ..... .... ........... 43

C. Sample System ............... ... ......... ......... ............. 48

V. RESULTS AND DISCUSSION .............................................. 53

A. Results .. . . . . . . ................ .......... . . .. . 53

B. Discussion ............................. 57

C. Outlook for the Future......................................... 59

REFERENCES ......................... 63

3

FIGURES

1. Oxygen Atom Number Density Profiles................................. 13

2. Schematic Representation of a tDMSP Satellite in Flight .............. 28

3. ESA Fields of View.. ..... ................................... . 29

4. ESA Degradation and Oxygen Atom Exposure as a Functionof Time ...................... 0..................0..................... 36

5. DMSP ESA Laboratory Experiment Apparatus ............................ 42

6. Microwave Discharge Atomic Beam Source ............................. 44

7. Mass Spectra of Oxygen Atom Beam .................................... 46

8. Pressure Dependence of 02 Dissociation .............................. 47

* 9. DMSP ESA Laboratory Experiment Optical System ....................... 50

10. Transmission Spectra ........................................... ..... 58

- It. High Energy Oxygen Atom Surface Chemistry Apparatus ................. 61

5

TABLES

1. Densities of Various Atmospheric Components......................... 12

2. Atmosphere Explorer C Monitored Wavelengths ......................... 15

3. DMSP Block 5D/I, TIROS-N, and NOAA-A SatelliteLaunch Dates and Degradation Extent............................... 31

4. 15.0 Pm Transmission of ESA Objective Lens Coatingwith Various Layers of Contamination ................................ 34

5. Emission Lines Observed from Microwave DischargeBeam Source Under Various Conditions ................ #............... 49

6. DHSP ESA Laboratory Experiment Conditions ........................... 54

7

I. INTRODUCTION

As the period for which satellite missions are required to function

increases, so the importance of the long-term effects of exposure of space-

craft materials to the upper atmosphere increases. The advent of the shuttle

orbiter, operating at relatively low altitudes, places further importance upon

the action of atmospheric species because of the higher atmospheric density

encountered in low Earth orbit. The interactions of atmospheric species, the

principal of which at orbital altitudes is atomic oxygen, with spacecraft

materials are poorly understood. This results, to no small degree, from

shortcomings in the experimental technology for simulating the orbital envi-

ronment in the laboratory.

A brief description of the atmospheric effects that are encountered by a

spacecraft in Earth orbit is provided in Section II. Some examples of satel-

lite systems, which have been or may be affected by atmospheric impact, are

presented. The present knowledge of the gas-surface interactions of atomic

oxygen and the state-of-the-art in experimental technology for simulating

atmospheric bombardment by oxygen atoms are reviewed.

The Defense Meteorological Satellite Program (DMSP) Earth Sensor Assembly

(ESA) offset anomaly is described in Section III. Evidence for linking this

misbehavior to atmospheric oxygen atom impact is presented. An experiment to

*test this link is outlined. The experimental apparatus is described in detail

in Section IV, and in Section V the results of the experimental program are

presented and discussed.

9

II. BACKGROUND

In this section a brief review of the atmospheric composition in low

Earth orbit is presented. Several satellite systems, which have been or may

be affected by atmospheric bombardment, are described. A review of the

effects of oxygen atom interactions with surfaces is presented, and the state-

of-the-art for production of atomic oxygen beams in the laboratory is

described.

A. ATMOSPHERIC COMPOSITION

The composition and density of the atmosphere above 120 km are functions

of several variables, such as local time, day of the year, geographic lati-

tude, sunspot activity, radio solar flux, and magnetic index. Neutral species

present at these altitudes include 0, N2, He, 02, Ar, and H. Molecular nitro-

gen is the dominant atmospheric neutral specie below 200 km, whereas atomic

oxygen dominates above 250 km. As one would expect, number densities of

neutral species decrease with increasing altitude. For example, at sunspot

maximum, the number density of oxygen atoms at 370 km is about 109 cm- 3, while

at 830 km it is about 107 cm- 3. In Table 1 are shown the densities of the

predominant atmospheric species at several altitudes. The dramatic variation

of oxygen atom concentration with solar activity is shown in Fig. 1. The num-

ber densities of the neutral elements and molecules in the atmosphere have

been modeled by Hedin.1- 3

Ions are also present above 60 km.4 The major ionic specie in the D

region (60 to 85 km) is NO+. In the E region (85 to 140 km), the most

abundant ions are NO+ and 02+. The NO+ ion is formed by reaction of 0+,I produced by solar x-rays, with N2. Nitrogen molecules are also photoionized,

but rapid dissociative recombination keeps the steady-state concentration ofN2

+ low. The F1 region lies between 140 and approximately 200 km. The pre-

dominant ions at its lower boundary are NO+ and 02+, whereas 0+ is the

principal ion near its upper boundary. The principal ion of the F2 region

(200 to 1300 km) is 0+, with N+ also present to a much lesser extent. At 830

km, the 0+ ion is approximately an order of magnitude lower in concentration

11

Table I. Atmospheric Composition (cm- 3 )

Altitude (km)

Speciesa 200b 3 0 0c 8 0 0d

N2 2.62 x 10 1.06 x 108 1.01 x 102

02 2.97 x 108 7.66 x 106 1.01

0 3.18 x 109 4.72 x 108 1.70 x 105

Ar 3.14 x 106 3.31 x 104 8.39 x 10- 5

He 5.16 x 106 3.11 x 106 4.28 x 105

H 1.92 x 105 1.63 x 105 9.88 x 104

NO+ 3 x 105 1.5 x 104

02+ 8 x 104 6 x 103

0+ 8 x 104 5 x 105 5 x 104

alon concentrations from Ref. 64bTemperature - 923 KcTemperature - 1004 KdTemperature - 1012 K

12

900

800 K

700 K~600

t-

500 - SUNSPOT MAXIMUM

STANDARD ATMOSPHERESUNSPOT MINIMUM

400

3002I510 210 3 e10 5106 1w108 0

OXYGEN ATOM NUMBER DENSITY (cm)

Fig. 1. Oxygen Atom Number Density Profiles

13

than oxygen atoms. Velocities of 0+ ions are about 1 km sec -1 . Above 1300

km, protons dominate. The ionosphere is electrically neutral, in the sense

that the positive ion density is equal to the electron and negative ion

density at all heights.

A spacecraft in low Earth orbit is bombarded by the atmosphere. A vessel

in orbit at 370 km, e.g., a shuttle sortie, with an assumed velocity of 8 km

sec- 1 is subjected to an oxygen atom flux of about 1014 cm- 2 sec -'. If all

these atoms were condensed upon the surface, a monolayer would form every 10

sec. At 830 km, e.g., DMSP, the flux is about 1011 cm- 2 sec - 1 , or a monolayer

in about 4 hr. Although the ambient atmospheric temperature is not high,

about 1000 K, the orbital velocity causes the atmospheric particles to strike

the spacecraft at high energies. The kinetic energies of 8 km sec -1 N2, O,

and He particles are approximately 9.3, 5.3, and 1.3 eV, respectively.

B. ATMOSPHERIC EFFECTS ON ORBIT

The atmosphere bombarding a spacecraft may interact with the surfaces of

the vessel in a number of ways which affect the surface properties. Among

these effects are: condensation, luminescence, sputtering, volatilization of

weakly bound surface deposits, and chemical reactions with the surface or with

impurities deposited thereon. Several satellite systems on which these

effects have been observed, predicted to be a problem, or postulated are

described herein.

1. ATMOSPHERE EXPLORER

Airglow measurements reported in 1977 performed by the Atmosphere

Explorer C (AE-C) spacecraft have suffered interference produced by an as yet

unexplained interaction with the neutral atmosphere at low altitudes. 5 This

interference appears as an increase in the measured brightness that is approx-

imately proportional to atmospheric density. The increase is greatest when

the instrument is looking in the direction of motion of the satellite, but the

effect is present at altitudes below 170 km even when the instrument is look-

ing upward. The individual wavelengths, which the AE-C experiments monitored

(as well as their origins), are listed in Table 2. Interference was present

at all wavelengths monitored. The intensity of the interference increases at

14

Table 2. Atmosphere Explorer C Monitored Wavelengths

X(A) Transition

3371 M 2(C 3 w + B )2 u g

4278 (B2 1+ x +)

5200 N(2D 45)

5577 0( Is + D)

6300 0( 1D 3p)

7319 0(2p 2D)

15

longer wavelengths. The effect is so severe that otherwise successful

observations coincident between AE-C and ground observatories had to be

discarded when AE measurements were made below 170 km.6 (AE-C was then in

elliptical orbit.)

Subsequent measurements made during periods of increased solar activity

show this effect to be present at altitudes as great as 400 km. Night-time

measurements of 7319 A emission confirm that the background is strongly

dependent on the direction of the sensor, emission being strongest when the

detector faces the atmospheric ram.6

When this effect was initially observed, it was hypothesized that the

emission arose from the NO2 continuum produced by reaction of atmospheric

nitric oxide with oxygen atoms adsorbed on the spacecraft. This supposition

has subsequently been proven incorrect. 6 Efforts are currently in progress at

Harvard and the University of Michigan to provide a satisfactory explanation

. of this phenomenon.

2. DISCOVERERS 26 AND 32

Since particles in the orbital atmosphere impact upon spacecraft surfaces

at high velocity, it is possible that surface material may be sputtered. Over

* a period of time, a surface of the space vehicle can be slowly eroded, result-

Ing in changes in various surface properties such as reflectivity or drag

coefficient.

Direct measurements of surface erosion have been made by McKeown and co-

workers 7- 9 using quartz crystal microbalances 10 orbiting on the Air Force

satellites Discoverers 26 and 32. Discoverer 26 was launched on 27 July 1961

into an elliptical orbit with apogee at 810 km and perigee at 228 km.

Discoverer 32 was launched on 13 October 1961 into a 404 by 232 km elliptical

orbit. The flights pass under the regions where protons and heavy ions

arising from radiation belts, solar flares, and the solar corona are

present. Thus any erosion of surfaces results from sputtering by molecules of

the upper atmosphere or by micrometeoroid impact.

164~ z

Gold surfaces normal to the velocity vector of the satellite, facing the

atmospheric ram, were observed to erode at a rate of 0.2 to 0.3 A/day. McKeown

has interpreted this erosion as sputtering by impact with atmospheric nitro-

gen. Nitrogen molecules strike the Discoverer spacecraft with a collision

energy of 9 eV. Sputtering rates were calculated to range from 1010 atoms

cu- 2 sec- 1 at 200 km to 107 atoms cm-2 sec- 1 at 800 km. These rates corre-

spond to a sputtering probability of 5 x 10-6 Au/N2

7 - 9

3. OGO-6

OGO-6 was launched on 5 June 1969 into a polar orbit with a perigee of

* 397 km and an apogee of 1098 km. Included on OGO-6 was an experiment to

determine the rate of deposition of satellite outgassing material upon clean

surfaces of the craft, as well as the rates at which the contaminated surfaces

were cleaned by desorption and sputtering.'1

The satellite outgassing flux consists of both volatile gases, such as

02, N2, CO2, and H20, as well as higher molecular weight contaminants, such as

outgassing, from adhesives and paints. The rates of spontaneous desorption

and atmospheric sputtering of contaminants were measured by observing mass

loss from contaminated surfaces after outgassing flux had dropped to a

negligible level. The mass loss caused by desorption, measured with the

experimental surfaces shielded from the atmospheric flux, was 1.2 x 10- 9 g-m- 2

sec'.

The mass loss rate for the combined processes of sputtering and desorp-

tion was measured at two values of angle of incidence of the atmosphere. At

normal incidence, the total mass loss was 3.5 x 10- 9 g-m- 2 sec -1 , which

implies a sputtering rate of 2.3 x 10- 9 g-m- 2 sec- 1. At incidence of 760 fromnormal, these rates were 3.8 x 10-9 and 2.6 x O-9 g-m 2 sec- .

These sputtering rates are not for the gold or aluminum surfaces them-

selves, but rather they are for the contaminants deposited thereon. Assuming

the contamination has an average molecular weight of 200, the probabilities of

an atmospheric molecule sputtering a contaminant molecule upon impact (P e)

are:''

17

Poo - 7 x 10- 5

P76° - 3 x 10

-4

The probability of sputtering is greater near grazing incidence because the

impacting molecule may- sputter a contaminant molecule without having to

reverse the direction of its momentum.

4. SATELLITE INFRARED EXPERIMENT (SIRE)

The objective of the SIRE program is to launch into orbit a highly

sensitive infrared telescope to survey targets and backgrounds. The detector

and its optics will be cryogenically cooled. (A NASA program, the Shuttle

Infrared Telescope Facility, is similar in nature.) Deposition of contam-

inants upon the cooled optics can degrade the performance of the sensor. The

sources of such condensates are the ambient atmosphere and the induced (out-

gassed) atmosphere of the satellite. It has been estimated that a cryodeposit

that is only I um thick will have a significantly deleterious effect.12

At 300 km, during periods of high solar activity, the density of atomic

oxygen is about 10 cm 3 . -3 Assuming a unit sticking coefficient to produce

solid 02, one calculates that an exposed surface normal to the velocity vector

of a spacecraft orbiting at 8 km sec- I will accumulate 02 at a rate of about

2 x 10-8 g-cm- 2 sec - ', which corresponds to about 50 um of deposit in only 1

hour. Such a rate of mass accretion is clearly unacceptable.

It has been suggested that an outward flow of helium or neon might pre-

vent contamination of the primary mirror of the SIRE telescope by directing

contaminants to other surfaces. 12'13 Approximate calculations of the effi-

ciency of such a scheme have been performed, but a calculation incorporating a

realistic model of each process involved in a purge flow to protect the tele-

scope mirror has not yet been performed. 13

5. DEFENSE METEOROLOGICAL SATELLITE PROGRAM (DMSP)

A degradation in performance of the ESA of the DMSP spacecraft and their

sister craft from the National Oceanographic and Atmospheric Administra-

18

tion/Television and Infrared Observation Satellite (NOAA/TIROS) Program has

been observed.14 Timing of the degradation indicates that an interaction with

atomic oxygen in the atmosphere may play a role in this degradation. This

problem is discussed in detail in Section IV.

C. OXYGEN ATOM SURFACE CHEMISTRY

Because atomic oxygen is the most abundant specie in the atmosphere from

200 to 1100 km, the gas-surface chemistry of atomic oxygen and techniques for

its investigation are discussed in the remainder of this section. A number of

different processes may occur when an oxygen atom interacts with a surface,

* such as elastic scattering, inelastic scattering, reactive scattering with the

surface or with some species deposited on the surface, occlusion (reaction

with the solid phase), absorption-desorption, recombination into 02, and

sputtering of loosely bound surface materials.

Because of the difficulty of producing moderately high energy (2 to

10 eV) oxygen atoms in the laboratory, knowledge of high energy oxygen atom

surface chemistry is severely limited. Indeed, a literature search produced

no indication that any investigations of oxygen atom interactions with sur-

faces in the 2 to 10 eV range have ever been conducted. A number of experi-

mental investigations of oxygen atom surface chemistry, at lower energies, are

described in succeeding paragraphs to indicate the state-of-the-art in this

area.

1. REACTIVE SCATTERING OF OXYGEN ATOMS FROM SURFACES

Madix and coworkers have investigated the oxidation of clean, single15

crystal surfaces of germanium and silicon by atomic and molecular oxygen.

In these experiments, atomic oxygen from an effusive beam source, whose most

probable velocity was less than 1 km sec- 1,16 was directed upon heated sur-

faces of silicon and germanium. From the experimentally determined flux of

oxygen atoms and from the observed rates of weight loss of the semiconductor

crystals, the reaction probabilities, a , were determined. These probabil-

ities were 0.15 to 0.5 for germanium surfaces between 800 and 1100 K, and 0.3

to 0.6 for silicon between 1300 to 1400 K. No dependence on surface tempera-

ture or wafer orientation was observed. These rates are an order of magnitude

19

greater than rates of oxidation for silicon and germanium by 020 The sole

products of these oxidation reactions have been observed to be the monoxides

of silicon and germanium. 14 Liu has reported a molecular beam, mass

spectrometric investigation of the reaction of atomic and molecular oxygen

with graphite surfaces.1 7 The source of atoms used in these experiments-1 161

produces a most probable velocity of about 1.8 km sec .16,17 These

experiments show that carbon monoxide is the dominant product of both atomic

and molecular oxygen reaction with 1000 to 1700 K surfaces. The reaction

probability for atomic oxygen under these conditions is 0.1 to 0.3, showing no

simple or large dependence upon substrate temperature. These probabilities

are greater than a factor of ten larger than those for molecular oxygen.

2. ELASTIC SCATTERING, INELASTIC SCATTERING, ANDATOM RECOMBINATION

Several measurements of the scattering of low energy (E < 0.33 eV) atomic

oxygen from surfaces have been reported. Many of these studies have been per-

formed with the goal of providing calibration data for mass spectrometers used

in upper atmospheric measurements. 18-25 These studies focused on atomic

oxygen losses on metal surfaces in mass spectrometer ion sources. Most of

these studies report only loss coefficients, i.e., whether an oxygen atom im-

pacting a surface will bounce off, or stick and recombine to form 02.

Wood 2 6 examined the rates of adsorption, occlusion, and recombination of

oxygen atoms on silver and gold surfaces. It was found that oxygen atoms

react on clean silver surfaces by adsorption and occlusion and on gold sur-

faces by recombination. These experiments showed that silver (at 300 K) takes

up atomic oxygen at a nearly constant rate well beyond monolayer coverage.

The nature of the occluded state was not determined. The occlusion rate was

measured to be 3.3 x 1014 atoms cm- 2 min -'. It was observed that gold adsorbs

up to a maximum of one monolayer of oxygen atoms at 300 K with a rate that

4 diminished with coverage.

Wood, Baker, and Wise 2 7 examined oxygen interactions with gold, silver,

titanium, stainless steel, and aluminum surfaces. Oxygen was shown to be lost

at gold surfaces by atom-adatom recombination. Occlusion was again shown to

be the primary loss mechanism on silver surfaces. Titanium reacted with

20

I

oxygen atoms to form a variety of solid oxides in a manner so complex that

kinetic measurements were not possible. Oxide-coated aluminum was virtually

inert to oxygen atom reaction or recombination. The interaction of oxygen

atoms with 302 stainless steel varied with pretreatment of the metal. In

general, loss rates were greater for steel than for noble metals. Under

certain conditions atomic oxygen reacted with carbon in the steel to form

27C02o2

Riley and Giese used atomic beam, mass spectrometric techniques to deter-

mine the probability of reflection of atomic oxygen from various surfaces,

such as Pyrex, stainless steel, Teflon, gold, and aluminum.2 8 They observed

reflection probabilities to lie between 45 and 60% for all these materials at

room temperature. The reflection probability for metals showed a strong, in-

verse temperature dependence dropping to below 10% at 673 K. (Such a depen-

dence upon surface temperature may cause one to doubt the accuracy with which

observations for these low energy collisions, < 0.33 eV, can be extrapolated

to describe events occurring at orbital velocity, about 8 km sec - 1, that is

about 5 eV.) The atoms which were not reflected in these experiments appar-

ently stuck to the surfaces. There was no evidence of recombination or pro-

duction of small molecules such as C02 or N20 in experiments carried at 10- 7

Torr.2 8 Hollister, Brackman, and Fite have described techniques for beam-

surface measurements of quantities of interest, such as reflection probabili-

ties and thermal accommodation rates.29

3. REMOVAL OF SURFACE CONTAMINANTS BY NEUTRAL ATOM BOMBARDMENT

Discharge flow experiments carried out at Boeing Aerospace Co. have shown

that ultraviolet polymerized contaminant films30 can, in some cases, be

removed by the action of radio-frequency excited oxygen.3 1- 3 3 Oxygen plasmas

successfully cleaned ultraviolet polymerized hydrocarbons from various mirror

and grating surfaces, silvered Teflon, and fused silica flats. Exposure to

oxygen plasma further degraded hydrocarbon-contaminated thermal control paint

samples. A synergistic effect is indicated by the fact that the oxygen plasma

had no affect upon pristine samples of the paint. Oxygen plasmas could not

remove silicon containing contaminants.

21

It was proposed that it was the oxygen atoms produced in the radio-

frequency discharge that caused the contaminant cleaning. However, further

experiments showed that a beam of radio-frequency excited 02 was successful in

cleaning hydrocarbon contaminants only when the contaminated surface was

exposed to the luminescent plume of a discharge source. This suggested that

ions, or some excited species contained in the luminescent plume, and not the

oxygen P atoms alone, were responsible for the cleaning.34 '3 5

Sputtering of adsorbed species by neutral atoms has been observed in

experiments on scattering of rare gas beams from a (111) silver

surface. 3 6 ,3 7 Surface cleanliness of epitaxially grown silver, which became

contaminated by hydrocarbon background gases in the molecular beam apparatus,

was monitored before and after bombardment by a high energy argon atomic

beam. A distinct threshold for contaminant sputtering of 1.0 to 2.0 eV

(dependent upon angle of incidence of the atomic beam) was observed. This

observed threshold for contaminant sputtering suggests that the failure to

clean contaminants in the oxygen atom beam/surface experiments described above

may owe to the collision energy in these experiments being below threshold.

The velocity distribution of the oxygen atom beam of References 34 and 35 was

not measured, but similar sources3 8 produce velocities of 2 km/sec or less, or

collision energies of less than 0.33 eV. One should note, however, that

*oxygen atoms impacting a satellite at orbital velocities have a collision

* energy of about 5 eV, which is well above the threshold observed for argon

sputtering of contaminant films.

5. EFFECT OF ATOMIC OXYGEN ON POLYMERS

Experiments in which radio-frequency excited oxygen impinged upon thin

films of polymers have shown that polymers can be rapidly oxidized by the

active species in discharged 02. Hansen et al. have studied oxidation of some

three dozen polymers by exposure to 1 Torr of radio-frequency excited 02•

Under their conditions the atomic oxygen concentration was of the order of

1014 to 1015 cm- 3 , with flow rates of 4 cm3 (STP)/min. All samples showed

measurable weight loss, typically on the order of 60 pg-cm m i . During

oxidation, emission of light fron excited 02 and oxygen was observed, as well

22

as emission from oxidation products, depending on the polymer under attack.

For example, CO2 and OH emission was observed when hydrocarbon polymers, such

as polyethylene were oxidized. SO bands, as well, were observed when sulfur-

containing polymers were treated.

Bulk properties of the polymers were not affected by the attack of atomic

oxygen, however, surface-dependent properties were affected. In many cases

simple ablation of the polymer surface occurred during oxidation by atomic

oxygen: polymeric material was simply oxidized and volatilized. Fluorinated

polymers behaved in this manner. In other polymers, especially polyethylene

and polypropylene, a hazy, oxidized surface layer was created. The oxidized

surface layer was hydrophilic. The surface oxide layer has the characteristic

of formation of strong adhesive bonds and provides sites for chemical attach-

ment of other molecules.4 0

D. EXPERIMENTAL TECHNOLOGY

In order to understand what physical and chemical effects may arise from

atmospheric oxygen atom bombardment of spacecraft surfaces, laboratory

investigations need to simulate that bombardment as closely as possible. In

particular, the velocity at which the atoms impact the surface ought to be-1

high, about 8 km sec * This high velocity of impact results, on orbit, from

the spacecraft sweeping rapidly through a more-or-less stationary atmos-

phere. Clearly, one may simulate this situation more readily in the

laboratory by producing rapidly moving gaseous atomic oxygen, which in turn

impacts upon a stationary surface. Such a simulation may most readily be

performed by means of atomic beam techniques.

Common atomic beam sources fall into two major classifications: effusive

and hydrodynamic (supersonic nozzle) flow sources.4 1 An effusive source con-

sists of a small chamber in which the beam material is contained. Atoms and

molecules leave the chamber by way of a small slit or orifice. The gas pres-

sure is controlled so that molecules and atoms leave by effusion rather than

bulk flow. The most probable velocity, vmp, of particles leaving such a

source is given by

23

3kT 1/2

where M is the particle mass, k is Boltzmann's constant, and T is the absolute

temperature. In order to obtain oxygen atoms at 8 km sec- I, such a source

must operate at over 40,000 K.

Higher velocities can be obtained by using hydrodynamic flow sources, in

which a gas at relatively high pressure undergoes an isentropic expansion

through a small nozzle. Relaxation during the gas expansion effectively

converts source enthalpy into a directed flow velocity, Vo,

Tp <M> V2 S C dT (2)0 o o Cp

where Ts is the source temperature, <M> is the average mass, and Cp is the

heat capacity at constant pressure of the gas. If the atomic oxygen were

diluted in helium, then <M> is approximately equal to that of helium, 4 g

mole-1 . Thus to obtain 8 km sec -1 oxygen atoms from a source of this nature

would require a Ts of 6000 K.

It is because of these high source temperatures required to generate 8 km

sec- 1 oxygen atoms that no one has yet reported in the literature such a

source. Numerous lower velocity sources of oxygen atoms have been des-

cribed. Nearly all rely on two basic techniques for producing oxygen atoms:

AC discharges of gaseous mixtures containing 02, or thermal dissociation of 02

in hot gas mixtures. Several sources that may be considered as archetypes are

described herein.

Gorry and Grice have used a 2450 MHz discharge to produce oxygen ators in

mixtures of 02 in inert gases. This source operates at relatively high pres-

sure, 90 to 200 Torr. At those pressures, flow from the source is hydrody-

namic (or very nearly so). Depending upon source pressure, the oxygen atomvelocity produced when He is the diluent peaks at 2.0 to 2.5 km sec - 1. Oxygen

atom fluxes are about 5 x 1017 sr- ' sec - 1.

I. [ Sibener et al. 3 8 a have described a source in which oxygen atoms are

produced in a radio-frequency (about 14 MHz) discharge. The source has been

24

operated at pressures above 200 Torr for mixtures of He and 02 producing

oxygen atom intensities ) 5 x 1018 sr- 1 sec-'. The velocity distribution of

this source peaked at 2 to 2.5 km sec - 1.

Liu 17 has reported a thermal dissociation source for the production of an

effusive beam of oxygen atoms. His source consists of a thorium oxide tube

which is wrapped with a tungsten filament. An AC current heats the filament,

which in turn heats the tube to temperatures in excess of 2100 K. Application

of kinetic theory 1 5 indicates that this source produces about 8 x 1014 sr- 1

sec- 1 of oxygen atoms, with a peak velocity of about 1.8 km sec - 1. Peplinski

has described a source of similar performance in which a source tube of

iridium is heated directly by an AC current.4 3

Extremely high atomic velocities have been produced in atomic beam

sources by the use of high power (several kilowatt) DC arcs to heat source

gases. Such devices have been successful in producing beams of rare

gases, 3 7 ,4 4 nitrogen atoms,4 5 and hydrogen atoms. 4 6'4 7 Velocities in excess

of 9.8 km sec - 1 for a binary mixture of Ar in He have been produced. However,

the introduction of 02 into such a discharge has resulted only in the rapid

oxidation of the arc electrodes. 4 8

A novel beam source in which a DC arc is used to heat a stream of He into

which 02 is injected (down stream of the arc) is under development. In this

manner, one hopes to obtain the high source temperatures of a DC arc while

avoiding the problem of electrode oxidation. Such a source under development

at the Air Force Geophysics Laboratory has produced 02 dissociation effi-

ciencies indicative of gas temperatures of 3500 K.49

25

III. DEFENSE METEOROLOGICAL SATELLITE EARTH

SENSOR ASSE4BLY OFFSET ANOMALY

The DMSP of the Air Force tries to maintain at least two satellites in

operation at all times in circular, sun-synchronous, polar orbits at an

altitude of 830 km. It provides U.S. military users with twice-daily global

and limited local, near-real-time cloud cover and Earth surface imagery and

information on atmospheric conditions.

In October 1978, one flight (F-3) of the DMSP Block 5D/1 lost Earth

stabilization because one of the space viewing detectors in its ESA was

showing an unexpectedly high count rate, a rate above the preset 900 count

limit. 14 When the ESA drifted above the 900 count limit, the spacecraft

control subsystem assumed that it had lost its lock on the Earth and initiated

an automatic Earth reacquisition cycle. After a brief stabilization, the

spacecraft again lost Earth stabilization. This sequence of events continued

until the detector limit was reset to a higher level, 1200 counts. Such an

effect has been observed on other DMSP Block 5D/1 flights and on similar

satellites.

A. EARTH SENSOR ASSEMBLY

The ESA is a static infrared horizon sensor that provides pitch and roll

data for the determination and control of spacecraft attitude (Fig. 2). These

data are used as a monitor of and backup for the primary attitude control sys-

tem on the DMSP satellites. The ESA is the primary attitude sensor for

NOAA/TIROS spacecraft. The operation of the ESA, in particular those elements

germane to the ESA offset anomaly, is described herein.5 0

The ESA has four detector arrays, which observe the Earth's limb, as

shown in Fig. 3. Each array consists of four thermopile detectors sharing a

common objective lens. Bandpass filters limit detector sensitivity to the 14

to 16 pm range. The detectors labeled A and B observe CO 2 V2 emission from

the atmosphere, whereas those labeled S provide a background reference. The

simple geometric relationship 50 between the A and B fields of view establishes

the horizon with respect to each of the four detector assemblies.

27

u

LU 4-4

041i

Lh-i

28u

DETECTOR ASSEMBLY DETECTOR ASSEMBLY1 4

-Y

Fig. 3. ESA Fields of View

29

The active junctions of the thermopile detectors viewing cold Earth or

space radiate heat and thus become colder than their reference junctions.

This produces a large negative background signal upon which small horizon

signals must be detected. This background negative signal is removed by

subtracting the output of the space viewing detector (S) from those of the

horizon viewing detectors (A and B). Because the responsivities of the detec-

tors are not perfectly matched, the magnitude of this correction is reduced by

using an offset radiation source (ORS) to compensate for heat lost to space by

the detectors. The ORS is a heated IRTRAN 2 ball, which emits poorly at

wavelengths shorter than 12 Pm. Radiation from the ORS is focused onto all

detector elements. The intensity of the ORS is regulated to drive the most

negative S detector to a predetermined value, near zero.

The four objective lenses of the ESA are also its windows. The f/1.7

germanium lenses are 59 mm in diameter. Their front surfaces are coated,

Optical Coating Laboratories, Inc. (OCLI), to reflect solar radiation from 0.4

to 2.4 Pm. The rear surface is coated to reflect solar radiation and block

transmission from 1.8 to 12 pm.

B. ESA DEGRADATION

Four satellites have been orbited in the DMSP series 5D/l. Launch dates

for these craft are shown in Table 3. Although the increase in count rate

(henceforth known as degradation) was not observed until flight 5D/1 F-3, it

has since been observed on all four satellites. 14 The same effect has been

observed on two NASA satellites TIROS-N and NOAA-A. (Those satellites perform

a function similar to that of DMSP. They fly circular orbits of 870 km,

slightly higher than DMSP.)

The greatest degradation rate for the 5D/I F-I and TIROS-N ESAs occurred

in February 1979. The 5D/I F-3 showed a maximum degradation rate in October

1978.14 Flight 2 also showed degradation, however, the exact timing of the

maximum degradation is unknown because of other problems. It appears,

however, that all degradations occurred within a few months of one another.5

(Degradation of 5D/LF-4 began immediately after launch.) The detector assem-

blies that were front facing, I.e., facing into the wind or ram direction

30

C-4

u0

@1 0 C 044 44 0

a 00414 * n e

04 tWoI 0 0

4

410

P,-

&0 4 0 4-. I-. cC r-

E0 (v to 0' -

4' w

00 0 - 0 4

0 1~

cc 0II~4J a

411 raw

'-4 + + I91

41 31

experienced the most degradation. However, there is some later evidence,5 1

which indicates that the rear detector assemblies also degraded, but to a

lesser extent (10 to 20% of the front facing detectors). Although degradation

was most immediately apparent in the S detectors, the offset occurs equally in

all four detectors of an affected quadrant. This fact was confirmed when

quadrant I of DKSP F-3 viewed space with all channels during an Earth requisi-

tion sequence.5 2

C. NIMBUS 6 AND 7 ERB INSTRUMENT

NIMBUS 6 was launched in June 1975 into an 1100 km high sun-synchronous,

polar orbit. NIMBUS 7 was launched in October 1978 into a 925 km high sun-

synchronous polar orbit. The Earth Radiation Budget (ERB) experiment was

designed to provide highly accurate radiation measurements of the sun and

.4 Earth. The measurements were designed to serve as a benchmark data set for

the long-term monitoring of the radiation of the global environment. Measure-

ments of radiation were made with 22 thermopile detectors with appropriate

filters. Ten of these spectral channels measured ultraviolet and visible

incoming solar radiation as the satellite orbited over the Antarctic. The

remaining channels measured Earth radiation in the 0.2 to 50 um region.5 3

The ERB instrument is mounted on the forward or ram side of the space-

* craft. Four solar channels (6,7,8, and 9) of NIMBUS 6 showed a slow continu-

ous drop in apparent solar radiance over 3 years. A drop in apparent solar

radiance (in channels 6 to 9) began on the first day of observations on NIMBUS

7. These decreases in detected solar radiance apparently result from contami-

nation of the solar channels by outgassing products of the spacecraft.14

The signal from the affected ERB channels of NIMBUS 6 began gradually to

increase in October 1978. Early in 1979 these signals increased rapidly. The

decay in signal on channels 6 to 9 of NIMBUS 7 reversed in January 1979, and

the signal then increased rapidly to a value greater than that observed whenmeasurements were begun. These affected channels faced into the atmospheric

wind; solar radiance observations made by rear facing detectors on NIMBUS 7

showed no such rapid fluctuations. This apparent rapid, uncharacteristic

32

clean up of contamination of the ERB instruments occurred during the same

period of time that degradation of the I4SP and NOAA/TIROS ESAs was occurring.

D. DEGRADATION MECHANISMS

There is strong evidence that the ESA offset is caused by a decrease in

the transmission of the objective lenses of the affected quadrants, which

reduced the radiative heat loss from the detectors. 5 2 The agent causing such

a decrease in transmission must affect the 14 to 16 um region of the spectrum

in which the ESA lens is transparent, but its effects need not be limited to

this region. Such an effect would also lower the apparent Earth radiance as

measured by the ESA. An analysis of the radiance measurements from quadrants

1 and 3 of DMSP F-3 before and after ESA degradation shows just such a reduc-

tion. The decrease in transmission necessary to explain both the detector

offsets and the reduced apparent radiance is of the order of 20%.52

Optical Coating Laboratories, Inc., has performed calculations to deter-

mine what alterations to the outer coating of the ESA objective lens are

necessary to account for a 20% decrease in transmission.54 Physical removal

of coating material as deep as 50% of the second layer does not radically

alter the transmission around 15 pm. The effects of deposition of films of

various thickness with an index of refraction of 1.6 and of various optical

absorption coefficients (8) have been calculated. The results of these compu-

tations are shown in Table 4.

Although these calculations by OCLI indicate that a contaminant layer of

great optical thickness in the long-wavelength infrared would be necessary to

produce the required 20% decrease in transmission, an induced opacity of the

lens remains the best explanation of the observed behavior of the ESA. That

the effect on all detectors of a degraded quadrant was the same tends to rule

out degradation of the detectors themselves or of optical elements unique to a

particular detector. A change in the ORS output alone would not account for a

decrease in the observed Earth radiance. Solar heating of the ESA is not

indicated since the offset appears both on ESA quadrants whose objective

lenses view the sun and on those which do not. Finally, electronic problems

in reading the detectors or communication are not indicated, inasmuch as the

effect is observed when either of two redundant sets of electronics is used.

33

00 n C-4 IsT 0r 00 0 0 It.-t 7% &' cc 1-. CV) C -4 0

ox .bC C) 0 c 0 0 0 0 0

00

as CV .).

CO 0 4 0b1 e

4-1400 0

to -4 414 a ' r s

0 9 *)*0~ 0o 0 C00

0' co-41t 4 .

05 0 4 0 c, Go C Ncx0

00

> )

0 X0f A I ~ % % C w dA

. 4j N N 1 7 , co r - O

414 0)

@3t

41

%C 00

E0 .5 m0 4 00 t04% en '.0 -4 co. n .

.0 Ci fs. i 4 c ' 0 n (%JA- C14 -C n ~ 4 0

343@

[4

That the greatest degradation rates for the DMSP and NOAA/TIROS ESAs

occurred simultaneously with the restoration of the NIMBUS solar channels

suggests that there may be a single cause for these phenomena. Any proposed

mechanism for these processes must explain this coincidence and explain why

the effects occur pred4ominantly on surfaces facing in the ram direction.

R. Predmore has considered several degradation mechanisms. 14 The data he

has accumulated indicate that because of the spacecraft geometries, the ESA

degradation could not be explained by condensation of spacecraft outgassing

q material alone. The effect of high energy charged particles (protons,

electrons, and alpha particles) has been ruled out because no increase in the

density of such particles occurred during the time of interest and because the

effect of such high energy particles would be omnidirectional. The possible

effects of the CAMEO experiment carried aboard NIMBUS 7, which released 38 kg

of lithium and barium at the end of October 1Q78, were also shown to be

negligible. 14

The anisotropy of the effects upon the ESAs and ERBs suggests the

possibility that interaction with atmospheric constituents may play a role in

the observed degradation. During the period in which rapid variations were

observed by the ESAs and ERBs there was an increase in solar activity, which

strongly affected the concentration of atomic oxygen and singly ionized oxygen

at the orbital altitude of the affected satellites. Predmore 14 reports a

calculation performed by Hedin, which gives the oxygen atom concentration

variation in 1978 and 1979 at an altitude of 800 km. The oxygen atom concent-

ration peaked in January and February 1979 and is reported to be the highest

concentration in over 20 years. A more recent calculation 51 by L. T.

Greenberg of the Optical Systems Department at Aerospace places the peak in

oxygen atom concentration somewhat later, in March 1979.

Assuming an orbital velocity of 8 km/sec, DMSP experiences bombardment by

an oxygen atom flux of about I012 cm- 2 sec -1 at sunspot maximum. Figure 4

shows a plot of the ESA degradation observed as a function of time. 14 Also

plotted is the intefrated oxygen atom flux. 5 1 The increase in the lower 0+

concentration takes the same form. In general, the failed detectors suffered

an oxygen atom integrated flux of about InlQ cm- 2 .

35

100-

0 OXYGEN ATOM EXPOSURE 150 0 ESA DEGRADATION

80- TIROS-N AND 5D Fl

-60- 10 ,9- E

00

0-0040 0z 0D

50

20-

0 10 20 30 40 50 60 70 80WEEKS

-)-1978 -- -- 1979

Fig. 4. ESA Degradation and Oxygen Atom Exposure as a Functionof Time

I

36I' ..

Although this collection of observations is by no means conclusive, they

suggest that the atmospheric fluxes of 0 and 0+ may play a role in the clean-

ing of the ERBs and the degradation of the DMSP, TIROS, and NOAA ESAs.

It is instructive to consider whether, in light of the OCLI calculations,

it is physically possible for 10 cm of oxygen atoms to produce a 20Z drop

in the transmission of the ESA objective lens. The density of solid oxygen,

1.43 g-cm- 3 , indicates the scale of the question at hand. If one assumes the

admittedly unrealistic situation that each atom of oxygen striking the DSP

ESA lens (which is at approximately 10°C) condenses to form solid oxygen, 1019

cm-2 of oxygen atoms would form a layer 1.9 Um thick. Some situations, which

are physically somewhat more reasonable, are examined herein.

One might assume that oxygen atom bombardment of the lens maintains an

adsorbed layer of oxygen atoms, which migrate into the lens material by

physical diffusion, thus altering the refractive index of the coating

material. The period over which degradation occurred on DMSP F-3 was

approximately 10 weeks. In order to produce a significant penetration of

oxygen atoms as deep as 1 pm into the coating, the diffusion constant for

oxygen atoms in the coating material would need to be of the order of i0-16

cm2 sec- I. This value is unreasonably large for a diffusion constant in a

solid at or about room temperature.5 5

Although diffusion of oxygen atoms into the surface is unlikely, reaction

of oxygen atoms with the surface material is conceivable. Calling the outer

coating material MX, one may propose a reaction scheme as follows:*

0 +MX +MX +0g s s ads

0Oad s + MX a + MO s+ ad0as sX s0 +Xads

X +0

S0 g g

ads + O 0g

*The identity of the outer coating material is known to the authors, however,

the manufacturer considers this information confidential.

37

The molar volume of MO is smaller than that of MX so it is reasonable to

expect that the oxide layer formed will be porous and, therefore, the rate

determining step may be the chemical process occurring at the interface. 56

This results in a rate determined by the exposed area of MX rather than by

diffusion. If the mechanism proposed above (which is substantially exothermic

at room temperature) were to occur at unit probability, a layer of MO approxi-

mately 1.2 uM thick could be generated.

A likely cause for the degradation of the ESA lens is the deposition of

some spacecraft outgassing material whose ill effects are exacerbated by the

incident atmosphere. 14'5' More than one role for the atmospheric flux is

possible. A return flux of material outgassing from the satellite could be

created by collisions with the incoming atmosphere. Impact of atmospheric

atoms and molecules could produce a new source of outgassing material, which

finds its way to the lens. Reaction of atmospheric oxygen with material on

the lens could alter the surface so as to enhance the probability of satellite

outgassing impurities sticking and remaining on the lens. Atmospheric oxygen

may react with already deposited contaminants or those being deposited to

produce a new species, which is more opaque in the 14 to 16 Im region.

Of these possible mechanisms, only the creation of a return flux by

atmospheric impact really admits to critical examination. If one assumes a

contaminant with a molecular weight of 200 and a specific gravity of 1.0,

a 1 Um layer would be comprised of about 3 x 1017 cm- 2 of molecules. If this

material were deposited over the period of degradation of the DMSP 5D/i F-3

ESA, the average flux of adhering contaminant molecules must have been 5 x

1010 cm- 2 sec - 1. A reasonable estimate of the ratio of contaminant flux

returned by atmospheric impact to the emitted flux of a spacecraft at the

altitude of DMSP is 10-5.57 Thus if the build up of a 1 um layer of contami-

nant by atmospherically induced return flux were the cause of the ESA degrada-

tion, the satellite must have been outgassing (under the assumptions of mole-

cular weight and density proposed above) at the unreasonably high3 8 rate-2 -1

of 1.7 jig-cm sec for 10 weeks.

38

E. DMSP ESA OFFSET ANOMALY LABORATORY EXPERIMENT

In order to test the hypothesis that atmospheric oxygen contributes to

the ESA degradation, a molecular beam experiment was undertaken to simulate

the neutral oxygen atom bombardment experienced by the DMSP spacecraft. The

experiment was designed to determine the effect of oxygen atom bombardment

upon the transmission of dielectric-coated germanium optics. This matter must

be addressed experimentally inasmuch as the available knowledge of oxygen atom

surface chemistry is insufficient to indicate what that effect might be.

The level of effort of this program and the time in which results were

sought dictated that we rely upon proven atomic beam source technology. Thus,

the impact velocity of the simulated oxygen atom bombardment of the satellite

optics was limited to less than 2.5 km sec - I , not the 6 to 8 km sec -'

experienced on orbit.

A positive result from such an experiment, that is, the observation of a

decrease in the transmission of the optic resulting from oxygen atom

bombardment, would strongly support the chemical degradation mechanism. A

completely successful experiment would allow one to determine the relative

importance of direct reaction with the surface and of reaction with some

satellite contaminant and would thus indicate areas of effort for preventing

such problems in the future. However, because the experimental bombardment

takes place at a velocity one third of that encountered on orbit, a negative

result would not conclusively disprove the hypothesis of atmospheric

degradation of the performance of the ootic.

The apparatus with which this experiment was performed is described in

detail in Section IV. The results of the program are presented and discussed

in Section V.

39

1

IV. EXPERIMENTAL APPARATUS

A schematic of the atomic beam apparatus used to test the chemical degra-

dation mechanism for the DMSP ESA is shown in Fig. 5. The major elements of

the facility are: a three chambered, differentially pumped vacuum system; a

source of oxygen atoms; a mass spectrometer to characterize the beam source; a

sample with a cryogenically cooled shroud tG protect it from spurious contami-

nation; and a light source and detector to monitor transmission of the sample

at or about 15 um. Each of these elements is described in more detail below.

A. VACUUM SYSTEM

The basic vacuum chamber consists of three stainless steel boxes, 18 in.

cubed. The first and third chambers are pumped by 10 in. oil diffusion pumps,

Consolidated Vacuum Corporation (CVC) and the second by a 6 in. oil diffusion

pump (CVC). Each pump is topped with a liquid nitrogen-cooled baffle (Mount

Vernon Research Co.). Each chamber may be isolated from its baffled pump by

means of an electropneumatically actuated, viton-sealed gate valve (Vacuum

Research Manufacturing Co.). The approximate pumping speeds of the 6 and 10

in. pump-baffle-valve stacks are 750 and 1500 1/sec, respectively. The pump-

ing fluid for chambers 1 and 2 is Dow Corning 704 silicon diffusion pump

oil. A hydrocarbon oil, Santovac 5, is used in chamber 3 to provide a cleaner

environment. The backing pump for chamber 1 is a 516 cfm Roots blower

(Penwalt Corp., Stokes Equipment Div. No. 1718). Chambers 2 and 3 are backed

in parallel by two mechanical pumps (Balzers DU090, 54 cfm and Sargeant-Welch

1397, 17.7 cfm).

Chamber 1, the source chamber, is ext.nded into chamber 2 and through the

top of chamber 2 (see Fig. 5). The purpose of this extension is to allow the

beam source to be mounted as near as possible to the third chamber in which

the bombardment occurs and thus to deliver the maximum available flux of

oxygen atoms to the surface. A nickel skimmer (Beam Dynamics, Inc., Model 2)

with an orifice diameter of 0.90 mm connects chambers I and 2.

41

I.Lac

ww

Ir-

0 0

0~

wd

44

Chamber 2 acts as a buffer to aid in the reduction of pressure between

the source and the experimental chamber. Also mounted in chamber 2 are a beam

chopper and a beam flag to aid in the characterization of the atomic beam (see

below). The orifice connecting chambers 2 and 3 is a 2.4 mm diameter hole

punched in a piece of stainless steel shim stock.

Chamber pressures are monitored by thermocouple gauges and by Bayard-

Alpert-type ionization gauges. With no gas flowing through the source, the

system pumps down to about 2.5 x 10- 7 Torr. Source operating conditions are

generally selected to keep the ionization gauge reading in chamber

1 1X 10- 3 Torr. Under these conditions the pressures in chambers 2 and 3

are about 1 x 10- 5 and 3 x 10- 7 Torr, respectively.

B. ATOMIC BEAM SOURCE

Oxygen atoms are produced by inducing a microwave discharge in gas mix-

tures containing 02, and allowing the discharged gases to expand into the

vacuum system to form a molecular beam. The source constructed for this

experiment uses a cavity very similar to that described in detail by Murphy

and Brophy.59 The cavity is a 3-1/4 wave foreshortened coaxial type with

microwave power coupling and cavity tuning of the Evenson design.6 0 (A recent

review6 1 of microwave-supported discharges by Zander and Hieftje provides

descriptions of various discharge cavities.) Power is supplied by a Raytheon

PGM1-10 2450 MHz magnetron power supply. The discharge tube is fused silica

with a 1-mm-diameter orifice. Figure 6 shows a schematic of the discharge

beam source.

Beam composition is measured with a quadrupole mass spectrometer

(Extranuclear Laboratories.) The major elements of the spectrometer are a

cross-beam, electron-impact ionizer including ion focusing optics, a

quadrupole mass filter with 17 mm diameter rods, and an 2lectron multiplier

(Johnson MM-I). The electron multiplier signal is fed to an Extranuclear

Laboratories 032-4 preamplifier whose output can be displayed on an

oscilloscope, digital voltmeter, X-Y recorder, or lock-in amplifier.

In order to discriminate mass spectrometer signals for oxygen in the

molecular beam from those for oxygen present as residual gas in the vacuum

43

wU0 z

z r 0

LL E_zj c

m c 0 coLU UJi~ Q

OO4Z~ <3 < < M cc

ow 0

0o o Cz

LUL

00

z

Ic

44mm uI

system, the phase sensitive detection method is used. A two-bladed rotary

chopper mounted in chamber 2 modulates the molecular beam at 47 Hz. A PAR

Model 186 lock-in amplifier is used to detect the modulated beam signal.

Figure 7 shows DC and AC mass spectra of the molecular beam.

The discharge source was operated with various gas compositions, source

pressures, and microwave power levels to determine those conditions most

suitable for performing the experiment. Conversion of oxygen into atoms was

determined from relative mass peak signals, IS/I0 , by the formula of Hiller

and Patch3 8b:

no 0 2 iO 1 (y L = i(3

n0 2 a0 n102

where P is the probability of dissociative ionization, (a0 /a ) is the ratio

of ionization cross sections, and n - I /1 with the discharge turned off.

Values of P and a0 /a are taken to be 0.3d and 1.30, respectively.3 8b The

effective percent d ssociation (ZD) is equal to loy/(y+2). Percent dissocia-

tions measured for pure 02, and for 4.88, 9.79, and 14.35% 02 in He all showed

a dependence on source pressure with a maximum dissociation of about 20 to

30%. When a small amount of water was added to the gas flow by bubbling the

feed gas through distilled, deionized water at 23 psi, the dissociation

increased dramatically, as has been reported in the literature,6 2 to a value

of 50 to 60% for all three He/0 2 mixtures. The percent dissociation increased

with microwave power coupled into the cavity. Figure 8 shows the dependence

of %D on source pressure for gas mixtures containing 02 in He.

It was determined that operating the source at 90 to 100 W with 18 to 20

Torr of 9.79% 02 in He, saturated with water at 23 psi at room temperature,

provided the best compromise between the requirement of maximum obtainable

beam intensity and the limitations on microwave power and chamber pressure.

Under these conditions the flux of oxygen atoms at the detector was estimated

by comparing height of mass peaks arising from the beam to those in the

residual gas to be 4 to 7 x 1014 cm- 2 sec - 1. The beam flux decreases with the

square of the distance from the source. Thus the flux of oxygen atoms

bombarding a sample is four times greater than that observed at the

45

(A) (B)I I%

C.

i",,

M

0 20 30 40 10204

z

1 -

/e (AMU) De (AMU)

Fig. 7. Mass Spectra of Oxygen Atom Beam. (a) DC spectrum,background gas; (b) DC spectrum, beam plus back-ground, discharge off; (c) AC spectrum, dischargeoff; and (d) AC spectrum, discharge on.

46

I

60

0-, 0.O

50 0

zo 40

i. C,00-

E 0-", 0A0 0

U. 20oLv o O2 AA 00o 0S0000 A A#

00 0 20 30

SOURCE PRESSURE (Torr)* 4.88% 02 - He

0 9.79% 02 - He

* 14.35% 02 - He

U 4.88% 02 - He WITH ADDED H20

9.79% 02 - He WITH ADDED H20

14.35% 02 - He WITH ADDED H20

0 02, NEAT

Fig. 8. Pressure Dependence of 02 Dissociation

47

detector. (See Fig. 5.) Therefore with these standard beam conditions, the1019 cm- 2 bombardment experienced by the IUISP spacecraft could be produced in

less than 2 hr. Somewhat longer bombardment times were used in practice

because of the approximate natures of the 1019 cm- 2 figure and of the source

flux calibration.

As well as producing atoms, the discharge emits light strongly. Emission

owing to H, He, OH and 0 were observed in the discharge. Table 5 lists the

various lines observed between 300 and 700 nm and their relative intensities

under three different source conditions.

C. SAMPLE SYSTEM

Samples bombarded were various dielectric-coated germanium disks (25 mm

o.d. x 1 mm thick) manufactured by OCLI. These samples were selected (except

in the case of the fina. experiment, see below) to have the same outermost

coating material as the ESA objective lens. The germanium flats were held

between viton O-rings in a 304 stainless steel block. The temperature of the

sample holder was controlled by two 10 W cartridge heaters (Hotwatt, Inc.),

which were regulated by an Omega Model 49 temperature controller. The sample

could be moved vertically (in and out of the beam path) and rotated about its

vertical axis, with the system evacuated.

The sample was surrounded by a 76 mm o.d. copper cylinder, closed at the

top and bottom, which was cooled to 77 K by liquid nitrogen. The purpose of

this shroud was to protect the sample from contamination by condensible resid-

ual gases. Two openings of 6 mm diameter allow the atomic beam to pass

through the cryoshroud. Two more openings at right angles to the beam path

provide a path for spectroscopic interrogation of the sample.

Transmission of the sample between 14 and 16 um was measured using a

glow-bar light source, collimated with a 51 mm diameter, 76 mm focal length

KCA lens. The detector was an Infrared Associates HCT 14-16 mercury-cadmium-

telluride photoconductive detector. The detector was protected with a 14 to

16 um band pass filter, OCLI 15000-9A. A 200 Hz Bulova tuning fork chopper

modulated the probing light, and the detector signal was monit.,red by a PAR

186 lock-in amplifier. Figure 9 shows a schematic representation of the

optical system.

48

4e

Table 5. Emission Lines Observed from Microwave Discharge

Beam Source Under Various Conditions

Intensity (arbitrary units)

Wavelength(nm) Assignment a : 02H20b 0 2c

668 He -p_ 1S 387

656 H Balmer a 26 165 5

654 d 2 1

588 He 3po - 3D 1000 14

505 He po lS 9

502 He iS -po 259

492 He Ipo I 1D 90

486 H Balmer F 5 26

476 d 144

471 He 3pq - 3S 47

447 He He3pO - 3D 267

439 He Ipo - ID 13

436 d 5

434 H Balmer y 4

414 He Ipo - D 2

412 He 3po _ 3S 6

403 He 3po - 3D 40

397 (H Balmer )e 1

396 He iS - po 22

389 He 3s- 3po 533

382 He 3po 3D 6

335 (He 3S - 3po)e 16

319 d 642 2300 -310 OH X -AE 18

a16 .5 Torr, discharge 85 WbStandard beam, see textc8 .5 Torr, discharge 100 WdUnidentifiedeAssignment uncertain

49

A. GLOWBAR LIGHT SOURCE

B. KC1 LENSC. IRIS

0. KCI WINDOWE. COPPER CRYOSHROUD (77 K)

F. SAMPLE (298 K)

G. VCM SOURCE (398 K)

S(a) H. ATOMIC BEAM AXISI. QUADRUPOLE MASS

SPECTROMETERJ. CHAMBER NO. 3

K. BANDPASS FILTER

_______D_ L. HgCdTe DETECTOR

M. FOCUSING MIRROR

N. ONE-THIRD METERMONOCHROMATER

Fig. 9. DMSP ESA Laboratory Experiment Optical System. (a) Spec-

tral dispersion and (b) total transmission.4

50

I

Volatile condensible materials (VCM), representative of satellite outgas-

sing impurities, could be deposited on the germanium flat by heating a sample

of the outgassing material, inside the cryoshroud. The VCM source was a brass

cylinder heated by six 10 W cartridge heaters (Hotwatt, Inc.) regulated by an

Omega Model 49 proportioning temperature control. The holder for the

outgassing material was inserted into the heated cylinder, secured by set

screws, and extended inside the cryoshroud. In all experiments reported here,

the outgassing material was GE RTV-615, polydimethylsiloxane, which had been

cured in air for 2 hr at 373 K. This contaminant was selected because

experimental data on a transmission spectra in the region of interest (in the

absence of oxygen atom bombardment) are available. 58 Furthermore because of

the heretofore mentioned inefficiency of oxygen plasma in removing silicon

containing surface contaminants, VCMs of this nature are considered to be a

likely source of difficulty. 31 During the deposition, the sample was raised

and rotated face-to-face with the VCM source. The exposed areas of the sample

and the outgassing material were of the same diameter, 19 mm, and were sepa-

rated by about 25 mm, thus approximately uniform coverage of the receiving

surface was provided. The typical outgassing period was 8 hr with the RTV

held at 373 K and the sample maintained at 298 K. After the deposition

period, the VCM source heaters were turned off and the source cooled rapidly

(by radiation to the cryoshroud).

51

V. RESULTS AND DISCUSSION

A. RESULTS

Seven separate experiments in which coated germanium flats were bombarded

with an atomic beam were performed. The natures of these experiments are

summarized in Table 6.

An initial experiment was performed to verify that the vacuum system was

sufficiently clean. An OCLI long-pass coated germanium filter with a cut-on

wavelength of 13.46 jm (L13460-9) was used. After it was mounted in the

sample holder, the transmission spectrum of the sample was measured using a

Perkin-Elmer 467 grating infrared spectrometer. The sample was then bombarded

for 5-1/2 hr with an oxygen atom beam produced under the standard conditions

described above. It was then removed from the system, in its holder, and its

transmission spectrum measured again. The 467 spectrometer measured the

transmission of the entire area of the disk while the atomic beam subtends

only a small central portion. Any change in transmission across the entire

area of the sample must result from spurious contamination from the vacuum

system. No change in the transmission spectrum of the disk indicative of such

contamination was observed.

The same sample was bombarded in the second experiment. This time the

transmission spectrum of the sample was between 14 and 16 um and was measured

in situ before and after bombardment, with the light source and detector

focused on the region of bombardment (see Fig. 9). Within the degree of

uncertainty of the measurement, no effect was observed after 5-1/2 hr of

bombardment. One must point out, however, that because of the long optical

path of this experimental configuration, the signal-to-noise ratio of this

measurement was poor. Thus the uncertainty in the measurement of the

transmission spectrum was so great that an effect would need to have been

dramatic to be observed.

The experimental procedure was altered to remedy this shortcoming. In

order to increase the precision in the transmission measurement, we measured

53

4 V 0

Do 4.4 z. 0 ..

0 0. 0. 404 0614 aia4 60Ai4

in 4. a!. 416 0 Or.A 09000C 0 0 060 0

"4C40 .4 .4 ~ 1 0441 0. 6 6 "4 w4 0. .44

.-4 m 000 co%a w4 10 9 w4 $40. a)14 v' 1 .4 - I 4 1fdV41

64 514 14. A. 0. S.11. 0. akto 601 0. x N " w m xK

co0 41 0 6 4' 41 to1 4)41 4 41

4 w a &A44 co

A34. 140 41C6 06

41'

.4 7

0

4~$ =*8* 14 -4 -4

01 = 0A 0=

i1 0 0m 0 0 0%

0 . 0% 0 0% Wo C o p

540 0 00 l- 0 000 00 0 0Q co0 - %0

r- M. 14 14:~ 0% 14 14 en 14

.*0 *. o -0

0to .4 w 6

-. 4 (a f U) W c -

I~ coIit

*- 6.4 14 4 14 4

'0~~ ~ u 4 0 04 ~* .O0 Z .. .2. .2. 0%0

10

54-

the transmission of the sample across the entire 14 to 16 trm band before and

after 6 hr of bombardment. The measured transmission of the OCLI 13460-9 disk

* was 79.4% before bombardment and 79.7% after bombardment. Since the uncer-

tainty in this measurement is of the order of 3 percentage points, the dif-

*ference between these two values is not significant.

In all three of the experiments described above, samples were bombarded

under the standard beam conditions in the absence of any added surface contam-

inant. In the fourth experiment the sample was coated with a VCH before

bombardment (Table 6). The sample used was a previously untreated OCLI

613460-9 filter; the VCM source was GE RTV-615. The transmission between 14

and 16 um (measured as in experiment 3) was 69.9% before VCM deposition, 72.9%

after VCM deposition, and 73.0% after 9 hr bombardment. Differences in these

numbers are not considered significant. Inspection of the sample after its

exposure to air revealed that because of misalignment of the apparatus during

VCH deposition, only half of the sample surface was coated with the impurity.

This experiment was, accordingly, repeated. The sample bombarded was a

clean OCLI-L08261-9 long pass filter with a cut-on wavelength of 8.26 urm.

Transmission of the filter was measured with the Perkin-Elmer 467 spectrometer

before installation in the apparatus. Transmission measured in situ over the

14 to 16 um band was 51.6% before VCM deposition, 50.4% after VCM deposition,

and 51.9% after 8.2 hr of bombardment. These differences are not considered

significant. Although no effect on the 14 to 16 Um transmission was observed,

there was a visible change in the VCM film over the area subtended by the

atomic beam. The area subjected to bombardment by the beam appeared to be

white, opaque to the eye, and rough, whereas unbombarded VCM films were smooth

and transparent. Measurement of the transmission spectrum of the sample

(after it was exposed to air) using the Perkin-Elmer 467 confirmed the absence

of any effect in the 14 to 16 pm band and showed only absorptions attributable

to the virgin VCM film.58 Thus the visibly altered area was not different to

the pure VCM in its transmission between 8.3 and 25 Pm. The substrate itself

was opaque at longer and shorter wavelengths, rendering further spectroscopic

measurements impossible.

55

Because the source produces species other than oxygen atoms, the

alteration to the VCH observed in experiment 5 cannot unambiguously be

ascribed to the action of atomic oxygen. The source also produces, as we have

noted above, light, and very probably metastable excited helium. In order to

attempt to discern the effect of oxygen atoms from those of other products of

* the source, two other experiments were performed.

An OCLI 13460-9 filter was treated with RTV-615 VCM and subjected for 10

hr to bombardment by the products of a discharge of pure helium. The 14 to

16 pm transmission of the sample was 75.5% before deposition, 67.4% after

deposition, and 74.5% after bombardment. The apparent reduction in transmis-

sion by the untreated VCM is difficult to explain inasmuch as it is at odds

with other experiments reported here and elsewhere.5 8 Although the 14 to

16 lm transmission of the sample was not different before treatment

and after bombardment, there was again a visible change in the appearance of

the VCM film. The area subtended by the beam source appeared darkened and

opaque to the eye. Because the sample is opaque at wavelengths shorter than

13.46 im, spectroscopic investigation of this visibly altered area was not

possible.

In the final experiment, an OCLI-6040005ST filter was used. This filter

was not selected to provide an outermost layer identical to that on the DHSP

ESA objective lense. This multilayer antireflection-coated germanium filter

is transparent to wavelengths longer than 1.8 Um, thus it permits more exten-

sive spectroscopic analysis of experimental results. Transmission of the

sample over the 14 to 16 Um band was 59.6% before VCM deposition, 58.5% after

VCM deposition (not significantly different), and 77.6% after 11-1/2 hr bom-

bardment by the beam produced by discharging 7.7 Torr of neat 02. (A longer

exposure time was used because the source operating on pure 02 produces a

lower absolute flux of oxygen atoms). The large change in transmissP , after

oxygen atom bombardment was surprising, especially in light of the fact that

the 14 to 16 im transmission of the sample after VCM deposition and bombard-

ment was greater than that for the clean sample. Inspection of the sample

after its exposure to air showed a transparent coating over the entire area

with a noticeable ring surrounding the area of bombardment (perhaps indicating

56

a difference in coating thickness in the area bombarded). Spectroscopic

examination of the sample with the P-E 467 tended to confirm that the bom-

barded area was more transparent in the LWIR than the unbombarded area

(Fig. 10). No sharp absorption features were observed other than those which

could be ascribed to virgin VCM on either area.

B. DISCUSSION

Experiments 1, 2, and 3 show that the LWIR transmission of these

dielectric-coated germanium optics is not affected by bombardment by oxygen

atoms at velocities less than 2.5 km sec -1 . This unambiguous result does not,

however, necessarily imply that oxygen atoms impinging on these surfaces at

orbital velocity will have no effect as well.

Experiments 4 through 7 imply that there is no dramatic effect on the

LWIR transmission VCMs produced from polydimethylsiloxane. Bombarded samples

showed no absorption features that differed from those of the untreated VCM.

Experiment 7 did show some change in the LWIR transmission of the sample.

However this alteration was manifest as an increase in the total transmission

at all wavelengths longer than 5 mm, upon which the same siloxane VCM

absorption bands appeared. No definite explanation of this result may be

offered here, but it is reasonable to suggest that this difference is a result

of a change in the thickness of the contaminant layer caused by bombardment.

(The appearance at the sample certainly supports this suggestion.) Extra

layers of different thickness will most probably alter the properties of

already complex multilayer coatings in different ways. It is not possible to

surmise from the data available how the impinging beam may have caused such an

alteration. Possible mechanisms are ablation by oxygen atoms, ablation by

excited 02 in the beam, and photolysis.

Although there were no clearly discernible effects to the bombarded

contaminant films in their LWIR transmission, effects resulting from the

bombardment were apparent (see above). Further spectroscopic investigations

of these affected surfaces were not possible because of the limited

transmission range of the substrates and the already complex reflectance

spectrum of the multilayer dielectric coatings of the germanium disks. The

57

V

100 -

80 -N

L 60

z~40I-

20

7 8 9 10 12 14 16 18 20 29

WAVELENGTH (micrometers)

Fig. 10. Transmission Spectra. Results of experiment 7 (dashed line,bombarded area; solid line area not bombarded, refer totext.)

58

Io.

appearance of the three bombarded contaminant films (cf experiments 5, 6, and

7) was different, but it is not possible unambiguously to ascribe these

differences to the variation in oxygen atom flux of the three bombarding

beams. In at least one qualitative way, opacity to the eye, the severity of

the effect on the film increased with increasing brightness of the beam source(Table

5).

That the results of this experimental program are in part ambiguous

points out the difficulty inherent in performing such measurements. These

experiments were required to elucidate the integrated effect of long exposure

* to the low orbit environment. In order to fulfill this role properly, an

experiment first must isolate an agent of that environment, in this case,

neutral oxygen atom bombardment. The experiment furthermore must simulate the

orbital experience sufficiently closely to ensure that any effect observed (or

not observed) can reasonably be expected to occur on orbit. Thus, a require-

ment of an integrated flux of 1023 m- 2 of oxygen atoms impinging at 6 to 8 km

sec - 1 is imposed. Technological limitations forced us to accept the

ambiguities introduced by using oxygen atoms of less than the desired velocity

and by the flux of ultraviolet light emanating from the discharge source.

C. OUTLOOK FOR THE FUTURE

The experimental investigation of the effect of atmospheric bombardment

of satellite surfaces by atomic oxygen is an ongoing effort in our labora-

tory. Facilities are currently under construction that will enable us to

model more realistically the atmospheric oxygen atom flux. A DC arc atomic

beam source, currently in its final stages of testing, has been built to

produce an intense source of atomic oxygen at velocities up to 8 km/sec. The

ultraviolet emission from this source will probably be more intense than that

produced by the microwave discharge sources used heretofore. A slotted disk

velocity selector6 3 will be used to filter out this radiation to determine

unambiguously the effects of oxygen atoms upon satellite surfaces. In

addition, the molecular beam apparatus used in the DMSP ESA anomaly studies

will be augmented by the addition of an ultrahigh vacuum chamber. Since this

chamber will be pumped by a high speed turbomolecular pump, it will provide

59

the ultraclean environment necessary for the proper performance of beam-

surface experiments. A schematic representation of the entire facility is

shown in Fig. II.

This facility will be used initially to measure the effect of high energy

oxygen atom bombardment on the reflectivity of front surface mirrors and on

the transmission of optical materials used in satellite sensors. These

measurements will be done on both clean and contaminated (with satellite

outgassed impurities) surfaces as a function of total oxygen atom flux and of

incident velocity. These experiments will provide an important extension of

the DMSP ESA study and will serve to resolve the ambiguities in the results

presented above. This facility will also allow investigation of other surface

phenomena which may be of importance in low orbit. Effects which could be

addressed (and the programs, which indicate their potential importance) are

sputtering (00-6), the volatilization of surface impurites (NIMBUS 6,7), and

chemiluminescent surface reactions (AE-C).

6

60

uJ

0

C)

44

C0)

00 A0

oo

bo

xLi

61

REFERENCES

1. A. E. Hedin, "Tables of Thermospheric Temperature, Density andComposition Derived from Satellite and Ground Based Measurements,"Vols. 1, 2, and 3, Laboratory of Planetary Atmospheres, GSFC,Greenbelt, Md. (January 1979).

2. A. E. Hedin, C. A. Reber, G. P. Newton, N. W. Spencer, H. C. Brinton,H. G. Mazer, and W. E. Pottes, "A Global Thermospheric Model Based onMass Spectrometer and Incoherent Scatter Data MSIS 2 Composition," J.Geophysical Res. 82, 2148 (1977).

3. A. E. Hedin, J. E. Salah, J. V. Evans, C. A. Reber, G. P. Newton, H. W.Spencer, D. C. Kayer, D. Alcoydi, P. Bauer, L. Cogger and J. P.McClure, "A Global Thermospheric Model Based on Mass Spectrometer andIncoherent Scatter Data 1, N2 Density and Temperature," J. Geophys.Res. 82, 2139 (1977).

4. F. S. Johnson, ed., Satellite Environment Handbook, Stanford UniversityPress, Stanford, Calif. (1961) pp. 27-39.

5. M. R. Torr, P. B. Hays, B. C. Kennedy, and J. C. G. Walker, Planet. andSpace Sci. 25, 173 (1977).

6. V. Abreu, University of Michigan, private communication.

7. D. McKeown, M. G. Fox, and J. J. Schmidt, "Measurement of SurfaceErosion from Discoverer 26," ARS Journal 32, 954 (1962).

8. D. McKeown, M. G. Fox, J. J. Schmidt, and D. Hopper, "Sputtering in theUpper Atmosphere," AIAA Journal 2, 400 (1964).

9. D. McKeown, "Surface Erosion in Space," Proceedings of the Third Inter-national Symposium on Rarefied Gas Dynamics, Paris (1962), p. 315.

10. D. McKeown, "New Method for Measuring Spattering in the Region NearThreshold," Rev. Sci. Instrum. 32, 133 (1961).

11. D. McKeown and W. E. Corbin, Jr., "Space Measurements of the Contami-nation of Surfaces by OGO-6 Outgassing and Their Cleaning by Sputteringand Desorption," NBS Special Publication No. 336, Space Simulation, J.C. Richmond, ed. (October 1970), p. 113.

12. J. P. Simpson and F. C. Witteborn, Appl. Opt. 16, 2051 (1977).

63

13. N. M. Harvey and R. R. Herm, Helium Purge Flow Prevention ofAtmospheric Contamination of the Cryogenically Cooled Optics of Orbit-ing Infrared Telescopes: Calculation of He-O Differential Cross Sec-tion, TR-0081(6432-03)-1, The Aerospace Corp., El Segundo Calif.7_5une 1981) (and references cited therein).

14. R. Predmore, "In-Orbit Degradation of Optics: TIROS/Block 5D ESA andNIMBUS ERB Sensors," NASA Goddard Memorandum (12 October 1969).

15. R. J. Madix and A. A. Sus, "Reactive Scattering of Atomic Oxygen FromClean Elemental Semiconductor Surfaces," Surf. Sci. 20, 277 (1970) (andreferences cited therein).

16. M. A. D. Fluendy and K. P. Lawley, Chemical Application of MolecularBeam Scattering, Chapman and Hall, London (1973), pp. 66-70.

17. G. N. K. Liu, "High Temperature Oxidation of Graphite by a DissociatedOxygen Beam," MIT DSR 72803, AFOSR, Massachusetts Institute ofTechnology Aerophysics Laboratory, Cambridge, Mass. (August 1973)(Contract F44620-71-C-0009).

18. H. B. Niemann, "An Atomic Oxygen Beam System for the Investigation ofMass Spectrometer Response in the Upper Atmosphere," Rev. Sci. Instrum.43, 1151 (1972).

19. A. E. Hedin, B. B. Hinton, G. A. Schmitt, "Role of Gas-Surface Inter-actions in the Reduction of Ogo 6 Neutral Particle Mass SpectrometerData," J. Geophys. Res. 78, 4651 (1973).

20. G. W. SJolander, "Atomic Oxygen-Metal Surface Studies as Applied toMass Spectrometer Measurements of Upper Planetary Atmospheres," J.Geophys. Res. 81, 3767 (1976).

21. L. R. Lake and A. 0. Nier, "Loss of Atomic Oxygen in Mass SpectrometerIon Sources," J. Geophys. Res. 78, 1645 (1973).

* 22. D. Offermann and K. U. Grossmann, "Thermospheric Density and Compo-sition as Determined by a Mass Spectrometer with Cryo Ion Source," J.Geophys. Res. 78, 8296 (1973).

I 23. H. B. Niemann and N. W. Spencer, "A Thermosphere Composition Measure-ment Using a Quadrupole Mass Spectometer with a Side Energy FocusingQuasi-Open Ion Source," J. Geophys. Res. 78, 2265 (1973).

24. L. R. Lake and K. Mauersberger, "Investigation of Atomic Oxygen in MassSpectrometer Ion Sources," Int. J. Mass Spectrom. Ion Physics 13, 425(1974).

64

25. 3. L. Hayden, A. 0. Nier, J. B. French, N. M. Reid, and R. J. Ducbett,"The Characteristics of an Open Source Mass Spectrometer Under Condi-tions Simulating Upper Atmosphere Flight," Int. J. Mass Spectrom. IonPhysics 15, 37 (1974).

26. B. J. Wood, "The Rate and Mechanism of Interaction of Oxygen Atoms andHydrogen Atoms with Silver and Gold," J. Phys. Chem. 75, 2188 (1971).

27. B. J. Wood, B. R. Baker, and H. Wise, Research Related to Measurementsof Atomic Species in Earth's Upper Atmosphere, Final Report ProjectPGU-6682, SRI (1970).

28. J. A. Riley and C. F. Giese, "Interaction of Atomic Oxygen with VariousSurfaces," J. Chem. Phys. 53, 146 (1970).

29. G. S. Hollister, R. T. Brackman, and W. L. Fite, "The Use of ModulatedAtomic Beam Techniques for the Study of Space-Flight Problems," Planet.Space Sci. 3 162 (1961).

30. R. B. Gillette and B. A. Kenyon, Appl. Opt., 10, 545 (1971).

31. R. B. Gillette, J. R. Hollchem, and George L. Carlson, J. Vac. Sci.Tech, 7 534, 1970.

32. R. B. Gillette, W. D. Beverly, and G. A. Cruz, Active CleaningTechnique for Removing Contaminants from Optical Surfaces in Space,Annual Report No. 1. The Boeing Company, Seattle,Wash. (March 1972)(Contract NAS 8-26385).

33. R. B. Gillette, and W. D. Beverly, Paper No. 71-463 presented at AIAA6th Thermophysics Conference, Tullahoma, Tenn. (April 1971).

34. R. L. Shannon and R. B. Gillette, Active Cleaning Technique Device,D180-18031-1, Boeing Aerospace Co., Seattle, Wash. (March 1973).

35. R. L. Shannon and R. B. Gillette, Laboratory Demonstration Model ActiveCleaning Technique Device, D180-18031-1, Boeing Aerospace Co., Seattle,Wash. (March 1974).

36. W. J. Hays, W. E. Rogers, and E. L. Knuth, J. Chem Phys. 45 2175(1972).

37. Shih-Min Liu, An Experimental Study of Interactions of HyperthermalAtomic Beams with (11) Silver Surfaces and Absorbed Molecules, ENG-7510, School of Engineering and Applied Science, UCLA, Los Angeles,Calif. (February 1975).

38. (a) S. J. Sibener, R. J. Buss. C. Y. Ng, and Y. T. Lee, Rev. Sci.Instrum. 51, 167 (1980).(b) D. H, Miller and D. F. Patch, Rev. Sci. Instrum. 40, 1566 (1969).

65

39. R. H. Hansen, J. V. Pascale, T. De Benedictis, and P. N. Rentzepis, J.Polymer Sci. A, 3 2205 (1965).

40. D. H. Reneker and L. H. Bulz, J. Macromol Sci-Chem AIO, 599 (1976).

41. M. A. D. Fluendy and K. P. Lawley, Op. Cit. pp. 65-87.

42. P. A. Gorry and R. Grice, "Microwave Discharge Source for theProduction of Supersonic Atom and Free Radical Beams," J. Phys. E. 12,857 (1979).

43. D. R. Peplinski, Crossed Beam Chemi-ionization Studies of Ba, Sr, Ca,La and Ce with Reactive Gases, M.S. Thesis, Ill. Inst. Tech. (1977).

44. W. S. Young, An Arc-Heated Ar-He Binary Supersonic Molecular Beam withEnergies up to 21 eV, Report 69-39, School of Engineering and AppliedSciences, UCLA, Los Angeles, Calif. (July 1969).

45. R. W. Bickes et al., "Utilization of an Arc Heated Jet for the Pro-duction of Supersonic Seeded Beams of Atomic Nitrogen," J. Chem. Phys.64, 3648 (1976).

46. K. R. Way, S. C. Young, and W. C. Stwalley, "A High Intensity Super-thermal Source of Hydrogen Atoms," ICPEAC IX. Abstracts of Papers,J. S. Risley and R. Geballe, eds., U. of Washington Press, Seattle,Wash. (1975).

47. K. R. Way, S. C. Young, and W. C. Stwalley, Rev. Sci. Instrum., 47,1049 (1976).

48. E. L. Knuth, UCLA, private communication.

49. J. A. Silver, Aerodyne Research Co., private communication.

50. Systems Analysis Report Vol. 4: Attitude Determination and Control,Radio Corporation of Amerca, Princeton, N.J. (30 January 1976) (Con-tract F04701-72-C-0221).

51. L. T. Greenberg, The Aerospace Corp., private communication.

52. K. Ward, The Relationship Between a Change in ESA Detector Offset and aChange in Measured Radiance, Barnes Project Report 2336, Barnes Engi-neering Co., Stamford Conn. (8 August 1979).

53. J. E. Sissala, ed., The Nimbus 6 User's Guide, Goddard Space Flight

Center, NASA, (February 1975).

54. G. Nilsen, Optical Coating Lab., Inc., private communication.

66

I

LII55. J. Askill, Tracer Diffusion Data for Metals Alloys and Simple Oxides,

IFI/Plenum, New York (1970).

56. K. J. Laidler, Chemical Kinetics, McGraw Hill, New York (1965), pp.316-318.

57. G. A. Bird, "Spacecraft Outgas Ambient Flow Interaction," AIAA J.Spacecraft, 18, (1), 31 (1981).

58. R. W. Phillips, L. U. Tolentino, and S. Feuerstein, SpacecraftContamination Under Simulated Orbital Environment, TR-0077(2270-30)-2,The Aerospace Corp., El Segundo, Calif. (6 July 1977).

59. E. J. Murphy and J. H. Brophy, Rev. Scl. Instrum. 50, 635 (1979).

60. F. C. Fehsenfeld, K. M. Evenson, and H. P. Broida, Rev. Sci. Instrum.36, 294 (1965).

61. A. T. Zander, and G. H. Hieftje, Appl. Spec. 35, 357 (1981).

62. P. M. Houpt, and G. H. W. Baalhuis, Appl. Spec. 34, 89 (1980).

63. J. L. Kinsey, Rev. Sci. Instrum. 37, 61 (1966).

64. J. Hofman, "Mass Spectrometer Measurements of Ionosphere Composition"in Recent Developments in Mass Spectroscopy," Proceedings of the Inter-national Conference on Mass Spectroscopy, Kyoto, Japan (1969), pp.1068-1072.

67

LABORATORY OPERATIONS

The Laboratory Operations of The Aerospace Corporation Is conducting exper-

imental and theoretical investigations necessary for the evaluation and applica-

tion of scientific advances to new military space systems. Versatility and

flexibility have been developed to a high degree by the laboratory personnel in

dealing with the many problems encountered in the nation's rapidly developing

space systems. Expertise in the latest scientific developments is vital to the

accomplishment of tasks related to these problems. The laboratories that con-

tribute to this research are:

Aerophysics Laboratory: Launch vehicle and reentry aerodynamics and heattransfer, propulsion chemistry and fluid mechanics, structural mechanics, flightdynamics; high-temperature thermomechanics, gas kinetics and radiation; researchin environmental chemistry and contamination; cw and pulsed chemical laserdevelopment including chemical kinetics, spectroscopy, optical resonators andbeam pointing, atmospheric propagation, laser effects and countermeasures.

Chemistry and Physics Laboratory: Atmospheric chemical reactions, atmo-spieric optics, light scattering, state-specific chemical reactions and radia-tion transport in rocket plumes, applied laser spectroscopy, laser chemistry,battery electrochemistry, space vacuum and radiation effects on materials, lu-brication and surface phenomena, thermlonic emission, photosensitive materialsand detectors, atomic frequency standards, and bloenvironmental research andmonitoring.

Electronics Research Laboratory: Microelectronics, GaAs low-noise andpower devices, semiconductor lasers, electromagnetic and optical propagationphenomena, quantum electronics, laser comunications, lidar, and electro-optics;communication sciences, applied electronics, semiconductor crystal and devicephysics, radiometric imaging; millimeter-wave and microwave technology.

Information Sciences Research Office: Program verification, program trans-lation, performance-sensitive system design, distributed architectures for

spaceborne computers, fault-tolerant computer systems, artificial intelligence,and microelectronics applications.

Materials Sciences Laboratory: Development of new materials: metal matrix

composites, polymers, and new forms of carbon; component failure analysis andreliability; fracture mechanics and stress corrosion; evaluation of materials inspace environment; materials performance in space transportation systems; anal-ysis of systems vulnerability and survivability in enemy-induced environments.

Space Sciences Laboratory: Atmospheric and ionospheric physics, radiation* from the atmosphere, density and composition of the upper atmosphere, aurorae

and airglow; magnetospheric physics, cosmic rays, generation and propagation ofplasma waves in the magnetosphere; solar physics, infrared astronomy; theeffects of nuclear explosions, magnetic storms, and solar activity on the

earth's atmosphere, ionosphere, and magnetosphere; the effects of optical,electromagnetic, and particulate radiations in space on space systems.